Complications of Colonoscopy
Total Page:16
File Type:pdf, Size:1020Kb
GUIDELINE Complications of colonoscopy This is one of a series of position statements discussing for adenomas larger than 1 cm is 12% to 17%.6-7 Although the use of GI endoscopy in common clinical situations. missed lesions are considered a poor outcome of colono- The Standards of Practice Committee of the American scopy, they are not a complication of the procedure per se Society for Gastrointestinal Endoscopy prepared this text. and will not be discussed further in this document. Com- In preparing this document, the authors performed a plications of bowel preparations are discussed in the search of the medical literature by using PubMed. Addi- American Society for Gastrointestinal Endoscopy Technol- tional references were obtained from the bibliographies of ogy Status Evaluation Report for Colonoscopy Prepara- the identified articles and from recommendations of ex- tion.8 pert consultants. When limited or no data existed from Over 85% of the serious colonoscopy complications are well-designed prospective trials, emphasis was given to reported in patients undergoing colonoscopy with results from large series and reports from recognized ex- polypectomy.3 An analysis of Canadian administrative perts. Position statements are based on a critical review of data, including over 97,000 colonoscopies, found that the available data and expert consensus at the time the polypectomy was associated with a 7-fold increase in the documents are drafted. Further controlled clinical studies risk of bleeding or perforation.9 However, complication may be needed to clarify aspects of this document, which data are often not stratified by whether or not polypec- may be revised as necessary to account for changes in tomy was performed. Therefore, complications of technology, new data, or other aspects of clinical practice. polypectomy are discussed with those of diagnostic This document is intended to be an educational device colonoscopy. A discussion of the diagnosis and manage- to provide information that may assist endoscopists in ment of all complications of colonoscopy is beyond the providing care to patients. This position statement is not a scope of this document, although general principles are rule and should not be construed as establishing a legal reviewed. standard of care or as encouraging, advocating, requir- ing, or discouraging any particular treatment. Clinical CARDIOPULMONARY COMPLICATIONS decisions in any particular case involve a complex anal- ysis of the patient’s condition and available courses of Cardiovascular and pulmonary complications related to action. Therefore, clinical considerations may lead an sedation are reviewed in detail in the 2008 American endoscopist to take a course of action that varies from this Society for Gastrointestinal Endoscopy Guideline for Se- position statement. This document is an update of the dation and Anesthesia in GI Endoscopy.10 Intraprocedural 2003 ASGE document entitled “Complications of colonos- cardiopulmonary complications have been variably de- copy.”1 fined to include events of unclear clinical significance, Colonoscopy is a commonly performed procedure for such as minor fluctuations in oxygen saturation or heart the diagnosis and treatment of a wide range of conditions rate, to significant complications including respiratory ar- and symptoms and for the screening and surveillance of rest, cardiac arrhythmias, myocardial infarction, and 11 colorectal neoplasia. Although up to 33% of patients report shock. In a study that used the Clinical Outcomes Re- at least one minor, transient GI symptom after colonos- search Initiative (CORI) database, cardiopulmonary com- copy,2 serious complications are uncommon. In a 2008 plications occurred in 0.9% of procedures and made up systematic review of 12 studies totaling 57,742 colonosco- 67% of the unplanned events during or after endoscopic 12 pies performed for average risk screening, the pooled procedures with sedation. Transient hypoxemia oc- overall serious adverse event rate was 2.8 per 1000 pro- curred in 230 per 100,000 colonoscopies, but prolonged cedures.3 The risk of some complications may be higher if hypoxemia was reported in only 0.78 per 100,000 colono- the colonoscopy is performed for an indication other than scopies. Hypotension occurred in 480 per 100,000 colono- screening.4 The colorectal cancer miss rate of colonoscopy scopies. CORI data may underestimate acute complica- has been reported to be as high as 6%,5 and the miss rate tions because of missing data and underreporting. A 2008 systematic review of randomized, controlled trials of pa- tients undergoing colonoscopy and/or EGD reported Copyright © 2011 by the American Society for Gastrointestinal Endoscopy much higher cardiopulmonary event rates with a weighted 0016-5107/$36.00 rate of 6% to 11% for hypoxemia and 5% to 7% for hypo- doi:10.1016/j.gie.2011.07.025 tension, depending on the specific drug regimen used.13 www.giejournal.org Volume 74, No. 4 : 2011 GASTROINTESTINAL ENDOSCOPY 745 Complications of colonoscopy In addition to acute complications, colonoscopy is as- perforation was 5 to 7 per 10,000 procedures (0.05%- sociated with an increased incidence of cardiovascular 0.07%) and not significantly different for procedures events in the 30-day postprocedure period. A study of coded as screening without polypectomy, diagnostic with- Medicare beneficiaries reported an unadjusted rate of car- out polypectomy, or with polypectomy (regardless of in- diovascular events requiring hospitalization or emergency dication).4 Finally, in a large study of 116,000 patients department visits of 1030 per 100,000 procedures, which undergoing colonoscopy at ambulatory endoscopy cen- was significantly higher compared with matched controls ters, there were 37 perforations (0.3%).21 (885/100,000 procedures).4 In a prospective study of pa- Surgical consultation should be obtained in all cases of tients undergoing colonoscopy at CORI sites, the event perforation. Although perforation often requires surgical rate at 30 days was 1.4 per 1000 for angina, myocardial repair, nonsurgical management may be appropriate in infarction, stroke, or transient ischemic attack.14 select individuals.22 There is an increasing number of case It is known that the risk of cardiopulmonary events reports demonstrating the feasibility of using endoscopic associated with colonoscopy is increased with advanced clipping devices to repair perforations.23 age,4 higher American Society of Anesthesiologists Physi- There is evidence that performance of colonoscopy by cal Status Classification System scores,15-16 and the presence an endoscopist with low procedure volume is associated of comorbidities.4 Appropriate assessment of anesthesia risk with increased risk of perforation and bleeding.9 Creating prior to colonoscopy may reduce cardiopulmonary compli- a fluid cushion at the base or under large polyps in order cations by ensuring that high-risk patients are co-managed to increase the degree of separation of the mucosal layers with other specialists (eg, cardiology, anesthesiology). Ap- has been described as a technique to potentially reduce propriate monitoring before, during, and after the procedure the risk of postpolypectomy perforation.24 It has been also may reduce the risk of complications. Unstable patients suggested that perforation rates greater than 1 in 500 for all should have non-emergent colonoscopy delayed as appro- colonoscopies or 1 in 1000 for screening colonoscopies priate. In addition, continuing aspirin and other antiplatelet should prompt evaluation of whether inappropriate prac- agents in the peri-endoscopic period may reduce the risk of tices are being used.24 cardiovascular events. The current American Society for Gas- trointestinal Endoscopy Guideline for Management of Anti- HEMORRHAGE thrombotic Agents for Endoscopic Procedures stresses that the risks of bleeding while receiving antithrombotic therapy Hemorrhage is most often associated with polypec- must be weighed against the risks of a thrombotic event if tomy, although it can occur during diagnostic colonos- that therapy is withheld.17 Although many thrombotic events copy. When associated with polypectomy, hemorrhage may be devastating, procedure-related GI bleeding is usually may occur immediately or can be delayed for several manageable and infrequently associated with significant weeks after the procedure.25 A number of large studies morbidity or mortality.17 have reported hemorrhage in 1 to 6 per 1000 colonosco- pies (0.1%-0.6%).2 A study analyzing over 50,000 colono- PERFORATION scopies by using Medicare claims found that the rate of GI hemorrhage was significantly different with or without Colonic perforation during colonoscopy may result polypectomy: 2.1 per 1000 procedures coded as screening from mechanical forces against the bowel wall, baro- without polypectomy and 3.7 per 1000 for procedures trauma, or as a direct result of therapeutic procedures. coded as diagnostic without polypectomy, compared with Early symptoms of perforation include persistent abdom- 8.7 per 1000 for any procedures with polypectomy.4 inal pain and abdominal distention. Later, patients may Polyp size has been reported as a risk factor for develop peritonitis. Plain radiographs of the chest and postpolypectomy bleeding in several studies.26-30 Addi- abdomen may demonstrate free air, although CT scans tional risk factors may include the number of polyps have been shown to be superior to the upright chest film.18 removed,31-32 recent warfarin therapy,28,33-34 and polyp